The aim of the proposed research is to determine if the syntactic comprehension dificits noted in both Broca's aphasics and conduction aphasics can be attributed to a short term memory deficit. Patients of both aphasic types will be tested on a wide range of tasks designed to assess aspects of short term memory relevant to syntactic processing. These tasks include tests of recognition versus recall, modality effects, the retention of item verus order information, and the types of memory codes which patients arae able to use. However, even if patients of one or both aphasic group show short term memory deficits plausibly related to syntactic comprehension deficits, it would still be possible that they suffered and additional impairment specifically in syntactic processing mechanisms. Thus, patients will also be tested on their ability to carry out syntactic analyses where the number of words they have to retain for the analysis is varied systematically. If patients can carry out syntactic analyses for strings within their memory span, but break down as the number of words increases, such will constitute strong evidence that their syntactic comprehension difficulties stem from a short term memory deficit. On the other hand, if the patients are impaired on all tests of syntactic comprehension independent of the number of words to be retained, a loss of syntactic processing mechanisms would be implied. The results of this study will have implications for theories of language and the brain as well as for the diagnosis and treatment of aphasic deficits. Since different areas of the brain are damaged in Broca's aphasics and conduction aphasics, any differences noted between these patient types in terms of either short term memory abilities or syntactic processing abilities will help to delineate functions of these different brain regions. The implications for patient treatment are more long range. However, the procedures developed in carrying out the proposed research may provide the first step toward the design of a battery of tasks that could efficiently assess whether a sentence comprehension deficit was due to a short term memory deficit. Since the treatment given a patient would differ depending on the underlying cause of a comprehension impairment, the results of the battery would guide the type of remediation offered.